Trajectories and patterns of productive activities on health in later life : a cross-national comparison study

Abstract

Background and Objectives: Since it emphasizes the productivity and contributions of older people to society, productive ageing has attracted great attention in gerontology. Although previous evidence illustrates the importance of promoting productive engagement in later life, most studies considered participation in productive activities as discrete behaviours and were focused on a single society. This thesis aims to expand knowledge of productive ageing by understanding three areas: first, the determinants of productive activities across countries; second, the reciprocal, longitudinal relationships between productive engagement and multidimensional domains of health; and finally, productive activity patterns and changes in these patterns as older people age. Methods: Data for three studies are derived from the Survey of Health, Ageing and Retirement (SHARE) and the English Longitudinal Study of Ageing (ELSA). Study 1 used multilevel logistic regression to examine the influences of individual-level factors (e.g., demographic characteristics and health status), family contact, and contact with friends on productive activities (paid work, volunteering, caregiving, grandparenting, and lifelong learning) over time (2011-2015) in England, France, and Denmark. Logistic regression was used to test the associations between community-level factors (e.g., I feel a part of this area) and productive engagement. In study 2, parallel process latent growth curve models were applied to examine directionality in the longitudinal relationships between productive engagement and mobility limitations, depressive symptoms, and episodic memory across countries. In study 3, latent class analysis (LCA) and latent transition analysis (LTA) were used to explore productive activity patterns and activity pattern changes as older people age. Results: For personal factors, in France, demographic characteristics, socioeconomic status, and health status were determinants of productive engagement; in England and Denmark, demographic characteristics, socioeconomic status, health behaviours, and health status were critical for productive engagement in later life. For social networks, family contact was a critical factor contributing to engaging in family care in all three countries; in England, contact with friends was associated with volunteer participation and, in France, with engagement in volunteer and lifelong learning activities. For community-level factors, community social environment was positively associated with volunteering in all three countries, while community physical environment was positively related to paid work in France. Study 2 found that productive engagement and physical, mental, and cognitive health declined significantly over time in all three countries. For physical health, in England and France, higher productive engagement was associated with a slower decline in physical health. Additionally, in England, individuals who reported higher productive engagement showed better cognition at baseline and a lower decline in cognition over time. No association between productive engagement and depression was found in all three countries. LCA in study 3 identified two classes of productive activity in England (volunteers and working-learning volunteers), France (family support and multiple engagers), and Denmark (low activity and high activity) in 2011. LTA revealed that productive activity patterns in later life were stable in England. In France, older people in multiple engagers had a relatively high probability (14%) of converting to family support. In Denmark, 75% of individuals converted from high activity to volunteering-family support, and 21% of individuals transitioned to the working-learning in 2015. Discussion and Implications: Personal factors do not always differentiate productive engagement behaviours, while social networks and community environment matter. The design of programmes and policies should consider these modifiable factors when facilitating productive engagement in later life. Efforts should be made to reduce barriers to productive engagement, considering that productive engagement brings a slower decline in physical and cognitive health. Furthermore, productive activities cluster in meaningful ways across countries, and activity profiles change as older people age. These characteristics may contribute to developing targeted policies and interventions for groups of varied activity patterns.